Africa is unfortunately popular for being riddled with Healthcare issues. Some of this is due to unfair branding, but there’s a lot of truth to the issue. Sub-Saharan Africa’s health expenditure as a percentage of GDP is just half the global average. In 2016, there were 8.8 million deaths in Africa—although about a third were from non-communicable diseases including stroke and ischemic heart disease, the majority due to communicable diseases including lower respiratory tract infections and HIV. We all remember the unfortunate Ebola outbreak of 2013-16 which took the lives of over 11,000 people in mainly Guinea, Liberia, and Sierra Leone.
Some of the key reasons for the high mortality rate from communicable diseases is the shortage of skilled health workers, ineffective systems, and the cost of care. Malawi, for example, has 1 doctor for every 60,000 patients. And Ghana has just about 34 radiologists, that is, 1 radiologist for every 800,000 people.
This is where artificial intelligence (AI) and data analytics comes into the picture. With AI, needed and unmet care like diagnostics can be automated to cut down cost and also empower the few available health workers to do more. Data analytics can be used to gather health data to accurately diagnose patients and provide precise treatment, as part of evidence-based healthcare. And also, it would allow Africa to more accurately monitor population health and track early symptoms of outbreaks. This can guide evidence-based policymaking.
African organizations like minoHealth AI Labs have been working on this. By using primarily deep learning and convolutional neural networks, they develop automated diagnostics and prognostics systems that can detect conditions including breast cancer, pneumonia, hernia and fibrosis from just x-ray images. And by leveraging data analytics, they are able to collect and analyze health data from facilities to better understand population and start tracking and forecasting outbreaks. With the assistance of local partners, including Christian Health Association of Ghana and National Catholic Health Service, they are testing these systems in Ghana.
Another African startup Ubenwa, has been using AI for cost-effective diagnosis of birth asphyxia in Nigeria by analyzing the sound of an infant cry. And Retina-AI has been developing AI apps for diagnosing and treating retinal diseases.
The effectiveness of AI in healthcare is being demonstrated globally. AI systems have been outperforming dermatologists in diagnosing skin cancer, they outperformed radiologists at diagnosing pneumonia in a Stanford study, and outperformed a pathologist in detecting breast cancer. Deep learning is being used to more accurately predict the survival of patients. By leveraging such a powerful technology in African healthcare, many tough challenges that have lingered for decades are now being solved.